Gondwana Research 36 (2016) 439–458
Contents lists available at ScienceDirect
Gondwana Research
j ourna l homepage: www.e lsev ie r .com/ locate /gr
Granites of the intracontinental termination of a magmatic arc: anexample from the Ediacaran Araçuaí orogen, southeastern Brazil
Leonardo Gonçalves a,⁎, Fernando F. Alkmim a, Antônio C. Pedrosa-Soares b, Ivo A. Dussin b,Claudio de M. Valeriano c, Cristiano Lana a, Mahyra Tedeschi b
a Departamento de Geologia, Escola de Minas, Universidade Federal de Ouro, Morro do Cruzeiro, 35400-000 Ouro Preto, MG, Brazilb Universidade Federal de Minas Gerais, Programa de Pós-Graduação em Geologia, CPMTC-IGC-UFMG, Av. Antônio Carlos 6627, Pampulha, 31270-901 Belo Horizonte, MG, Brazilc TEKTOS, Geotectonics Study Group, Universidade do Estado do Rio de Janeiro, Rua São Francisco Xavier 524/4006-A, 20559-900 Rio de Janeiro, RJ, Brazil
⁎ Corresponding author. Tel.: +55 31 87101840.E-mail addresses: [email protected], leon
(L. Gonçalves), [email protected] (F.F. Alkmim), pedro(A.C. Pedrosa-Soares), [email protected] (I.A. Dussin(C.M. Valeriano), [email protected] (C. Lana), m(M. Tedeschi).
http://dx.doi.org/10.1016/j.gr.2015.07.0151342-937X/© 2015 International Association for Gondwa
a b s t r a c t
a r t i c l e i n f oArticle history:Received 28 January 2015Received in revised form 7 July 2015Accepted 29 July 2015Available online 29 August 2015
Handling Editor: A.S. Collins
Keywords:GranitesMagmatic arc terminationEdiacaranRio Doce arcAraçuaí orogen
The Araçuaí orogen of southeastern Brazil together with the West Congo belt of central West Africa form theAraçuaí–West Congo orogen generated during closure of a terminal segment of the Neoproterozoic AdamastorOcean. Corresponding to an embayment in the São Francisco–Congo Craton, this portion of the Adamastor wasonly partially floored by oceanic crust. The convergence of its margins led to the development of the Rio Docemagmatic arc between 630 Ma and 580 Ma. The Rio Doce magmatic arc terminates in the northern portion ofthe Araçuaí orogen. Granitic plutons exposed in the northern extremity of the arc provide a rare opportunityto studymagmatism at arc terminations, and to understand the interplay between calc-alkaline magma produc-tion and crustal recycling. The plutons forming the terminus of the arc consist of granodiorites, tonalites andmonzogranites similar to a magnesian, slightly peraluminous, calcic- (68%) to calc-alkaline (24%), with minoralkali-calcic (8%) facies, medium- to high-K magmatic series. Although marked by negative Nb–Ta, Sr and Tianomalies, typically associated with subduction-relatedmagmas, the combined Sr, Nd and Hf isotopic data char-acterize a crustal signature related to anatexis of metamorphosed igneous and sedimentary rocks, rather thanfractional crystallization of mantle-derived magmas. Zircon U–Pb ages characterizes two groups of granitoids.The older group, crystallized between 630 and 590 Ma, experienced a migmatization event at ca. 585 Ma. Theyounger granitoids, emplaced between 570 and 590 Ma, do not show any evidence for migmatization. Most ofthe investigated samples show good correlation with the experimental compositional field of amphibolitedehydration-melting, with some samples plotting into the field of greywacke dehydration-melting. The studiedrocks are not typical I-type or S-type granites, being particularly similar to transitional I/S-type granitoids de-scribed in the Ordovician Famatinian arc (NW Argentina). We suggest a hybrid model involving dehydration-melting of meta-igneous (amphibolites) and metasedimentary (greywackes) rocks for magma production inthe northern termination of the Rio Doce arc. The real contribution of each end-member is, however, a challeng-ing work still to be done.
© 2015 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.
1. Introduction
Major and trace element data, together with Lu–Hf in zircon, whole-rock Sm–Nd and Rb–Sr isotopes provide sensitive discriminators of tec-tonic and/or magmatic processes operating above subduction zones andwithin collisional orogens (Rudnick, 1995; Kemp and Hawkesworth,2003; Kemp et al., 2006; Liu et al., 2013; Niu et al., 2013). These chemistryand isotope datasets can also be used to discriminate between a range of
[email protected]@pq.cnpq.br), [email protected]@gmail.com
na Research. Published by Elsevier B.
geological processes involved in the pre-collisional to collisional stages ofan orogenic belt evolution, such as crustal reworking, crust–mantle inter-actions, and production of juvenile magmas from the mantle.
The Araçuaí orogen of southeastern Brazil and the West Congo beltof southwesternAfrica once lie in the central portion ofWestGondwana(Alkmim et al., 2001). Together, they form the Araçuaí–West Congoorogen (AWCO), generated during closure of a terminal branch of theNeoproterozoic Adamastor Ocean (Pedrosa-Soares et al., 1992, 2001;Brito Neves et al., 1999; Cordani et al., 2003; Alkmim et al., 2006). Em-placement of ophiolite slivers, development of intra-oceanic andcontinental-marginmagmatic arcs, and collision of the plates represent-ed by the São Francisco–Congo, Paranapanema, Rio de la Plata andKalahari cratons record the Adamastor Ocean closure from theCryogenian to Ediacaran periods (e.g., Pedrosa-Soares et al., 1998,2011; Campos-Neto, 2000; Alkmim et al., 2001, 2006; Basei et al.,
V. All rights reserved.
Fig. 1. A) The Araçuaí–West Congo orogen and related cratons in the context ofWest Gondwana (after Alkmim et al., 2006; Noce et al., 2007). Dark gray= cratons; light gray= orogenicbelts. B) Relative positions of the Neoproterozoic orogenic belts (in gray) presently exposed along the South American and African margins of the Atlantic (modified from Porada, 1989).Redpolygon and regions indicate, respectively, the approximate location of Fig. 1C andplutons forming the RioDoce and RioNegro arcs. C) Simplified geologicalmapof theAraçuaí orogen,with a box indicating the studied region (modified from Pedrosa-Soares et al., 2008). ACSZ = Abre Campo Shear Zone; SFC = São Francisco Craton. Cities: BH= Belo Horizonte; GV =Governador Valadares; TO = Teófilo Otoni; AF = Águas Formosas; RP = Rio do Prado.
440 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
441L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
2010; Tupinambá et al., 2012; Heilbron et al., 2013). Representing thenorthern segment of the Adamastoran orogenic system (e.g., Goscombeand Gray, 2008), the AWCO exhibits a very peculiar configuration(Fig. 1), as it ended against a continental bridge linking the basementcounterparts located in Bahia state (eastern Brazil) and Gabon (WestAfrica), the São Francisco–Congo cratonic bridge (Torquato andCordani, 1981; Porada, 1989). In spite of this unique tectonic setting,the AWCO exhibits Late Cryogenian to Ediacaran ophiolite slivers andthe Ediacaran Rio Doce magmatic arc, suggesting that its precursorwas a gulf-like inland-sea embayment partially floored by oceaniccrust (e.g., Pedrosa-Soares et al., 1998, 2011; Nalini et al., 2000; Alkmimet al., 2006; Queiroga et al., 2007).
The exhumed crystalline core of the Araçuaí orogen exposes awholeseries of calc-alkaline plutons dated between 630 Ma and 580 Ma andgrouped in the G1 Supersuite (Figs. 1 and 2), which represents the plu-tonic part of the Rio Doce magmatic arc (Figueiredo and Campos Neto,1993; Nalini et al., 2000; Pedrosa-Soares and Wiedemann-Leonardos,2000; Pedrosa-Soares et al., 2011; Gonçalves et al., 2014).
Fig. 2. Simplifiedgeologicalmap of the study area showing the sample sites for petrographic, geoRancho Alegre (samples LG28, LG49 and LG63), Fz = Felizburgo, BJV= Bom Jesus da Vitória (s(sample LG21),W=Wolff (sample LG161), C=Caraí (sample LG12) and Fazenda Liberdade (sfor further details.
Magmatic arcs developed along active margins rarely died out inintracontinental domains. Because of the peculiar tectonic setting ofthe AWCO, the G1 Supersuite plutons exposed in the northern Araçuaíorogen apparently mark the termination of the Rio Doce arc in a conti-nental setting (Figs. 1 and 2). What is the chemical signature of granitesformed in a magmatic arc dying out in an intracontinental domain? Arethey chemically similar to granites formed in arcs developed on activeplate margins or more akin to collisional granites? What is the extentof crustal–mantle interaction and crustal recycling in such a highlyensialic setting? The G1 plutons of the northern Araçuaí orogen consti-tute a unique natural laboratory for providing answers to these ques-tions and investigating features and process not yet well documentedin the literature. In this paper, we present results of a detailed field, pet-rological, geochemical and geochronological study performed on G1Supersuite plutons exposed in the northern end of the Rio Doce arc. Inorder to discuss the nature of the granitic rocks generated in this partic-ular setting, we compare our results with geochemical datasets fromother segments of the Rio Doce arc and arc-related granites worldwide.
chemical, isotopic and geochronological analyses. The studied anddatedplutons are: RA=ample LG31), PS= Pedra do Sino (samples LG27 and LG173), SV= São Vítor, T = Topázioample LG29). For simplicity the samples are shownon themap just with numbers. See text
442 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
2. Geological setting
The area we selected for study, encompassing a segment of ca.10,000 km2 of the crystalline core of the Araçuaí orogen, is located be-tween 16°30′ and 18°00′ S Lat., and 40°30′ and 42°00′ W Long.,i.e., between the towns of Teófilo Otoni and Rio do Prado in northeast-ern Minas Gerais State, Brazil (Figs. 1 and 2). We chose this region be-cause it is entirely covered by geologic maps (Fontes, 2000; Moreira,2000; Paes, 2000; Sampaio, 2000; Gomes, 2008; Paes et al., 2010) andcontains good exposures of the northernmost plutons of the G1Supersuite (Fig. 2).
The evolution of the Araçuaí orogen between the Early Ediacaran upto the Cambrian–Ordovician boundary produced an enormous volumeof granitoids (including their Opx-bearing equivalents) and subordinatemafic rocks. Using field relationships, structural features, chemical andgeochronological data, these igneous rocks were grouped into five as-semblages, the G1, G2, G3, G4 and G5 supersuites (Pedrosa-Soareset al., 2011).
The granitic rocks forming the G1 Supersuite in the study area in-trude with diffuse and sharp contacts the metasedimentary rocks ofthe Jequitinhonha Complex and the meta-volcanosedimentary RioDoce Group, and are in turn cut by granites of the G2, G3 and G5supersuites (Fig. 2).
Representing the oldest rock assemblage in the study area, the Jequi-tinhonha Complex (Fig. 2) consists of garnet–biotite gneisses, kinzigites,quartzites, graphite-rich gneisses, marbles, and metamafic rocks(Almeida and Litwinski, 1984; Paes et al., 2010; Gonçalves-Dias et al.,2011). U–Pb determinations on detrital zircon grains from a quartziteintercalated with the paragneisses have yielded a maximum deposi-tional age of 898 ± 8 Ma for protoliths of the complex (Gonçalves-Dias et al., 2011), which are interpreted as high grade equivalents ofthe Araçuaí orogen type unit, the Tonian–Cryogenian Macaúbas Group(Pedrosa-Soares et al., 2008; Gonçalves-Dias et al., 2011, and referencestherein).
TheRioDoceGroup crops out in the southwesternportion of the studyarea (Fig. 2), and comprises pelitic schists, quartzites, paragneisses, andmetagreywackes. The group also contains dacitic and rhyolitic rocksdated around 585 Ma (Vieira, 2007; Gonçalves et al., 2010, 2014). Themaximum depositional age of the group, estimated at 594 ± 3 Ma, wasconstrained by the ages of detrital zircons extracted from schists andmetagreywackes (Vieira, 2007). Field observations and petrographicstudies led Vieira (2007), Pedrosa-Soares et al. (2011), and Gonçalveset al. (2010, 2014) to interpret the volcanic rocks of this unit as equivalentto the G1 rocks and the group as a whole as a Rio Doce arc coversuccession.
In the study area, the G1 Supersuite includes small plutons (up to5 km2) and stocks (up to 80 km2). The main bodies, known as RanchoAlegre, Felizburgo, Pedra do Sino, Bom Jesus da Vitória, São Vítor andTopázio (Fig. 2), consist of tonalites (including Opx-bearing terms),granodiorites, and minor monzogranites. They are essentiallyperaluminous rocks (the ASI average is 1.07), involving calcic (68%),calc-alkaline (24%) and alkali-calcic (8%) facies, occasionally containingtitanite or garnet, but commonly without hornblende. These rocks areisotropic to foliated, with local migmatitic features, and host bothmeta-igneous and metasedimentary enclaves, being the latter by farthe more frequent (e.g., Paes et al., 2010 and references therein).
In the remaining portions of the Araçuaí orogen, the G1 Supersuiteconsists of medium- to high-K, calc-alkaline, tonalitic to granitic bodies(including their Opx-bearing equivalents), and subordinate mafic to in-termediate plutons containing gabbroic to enderbitic facies (Novo et al.,2010; Tedeschi, 2013; Gonçalves et al., 2014). Typical for themajority ofthe G1 rocks is a large population of mafic enclaves (Nalini et al., 2000;Gonçalves et al., 2010, 2014; Pedrosa-Soares et al., 2011). All these char-acteristics, together with the pre-collisional nature of the G1 Supersuite,ledmost authors to interpret this rock assemblage as representative of acontinental magmatic arc (e.g., Nalini et al., 2000, 2005; Pedrosa-Soares
and Wiedemann-Leonardos, 2000; Pedrosa-Soares et al., 2001, 2008,2011; Martins et al., 2004; Gonçalves et al., 2010, 2014; Novo et al.,2010; Tedeschi, 2013), named Rio Doce arc by Figueiredo and CamposNeto (1993). The development of the Rio Doce arc, constrained by zir-con U–Pb ages between 630 and 580 Ma, is viewed as a consequenceof the consumption of the oceanic segment of the Adamastor terminalbranch, which took place before the collisional event that led to thefull development of the AWCO (Pedrosa-Soares et al., 1998, 2001,2008; Alkmim et al., 2006; Gonçalves et al., 2014).
The G2 Supersuite occurs in the form of isolated plutons or batho-liths mainly in the central and southern portions of the study area(Fig. 2), where foliated biotite–cordierite–garnet–sillimanite granitespredominate (Gradim et al., 2014). Made up essentially of S-type gran-ites, the G2 Supersuite show U–Pb ages between 590 Ma and 545 Ma(e.g., Nalini et al., 2000; Pedrosa-Soares and Wiedemann-Leonardos,2000; Pedrosa-Soares et al., 2011; Gradim et al., 2014). This interval cor-responds to themain phase of deformation andmetamorphism record-ed in the Araçuaí orogen (Silva et al., 2002, 2011; Alkmim et al., 2006;Pedrosa-Soares et al., 2011; Gonçalves et al., 2014; Gradim et al., 2014).
Commonly associated with the G2 rocks, the G3 Supersuite is madeup of isotropic garnet–cordierite–sillimanite leucogranites. Mostlycomposed of S-type granites, the G3 rocks are not individualized inthe study area, due to limitations on the scale of mapping. Its magmaticcrystallization is constrained by zircon ages between 545 Ma and530 Ma, representing the late-collisional stage of the Araçuaí orogen(Pedrosa-Soares et al., 2011; Gradim et al., 2014).
The G4 Supersuite occurs as zoned plutons or batholiths in the re-gion to the south of the study area (Fig. 1) and consists of non-foliatedbiotite granite, two-mica and muscovite–garnet leucogranites. Com-monly, G4 plutons show igneous flow structures and are peraluminous.Zircon U–Pb data constrain the magmatic crystallization age of the S-type G4 granitoids between 530 Ma and 500 Ma (e.g., Pedrosa-Soareset al., 2011 and references therein). According to Alkmim et al. (2006)and Pedrosa-Soares et al. (2011), together with the G5 Supersuite, theG4 granites represent the product of a magmatic event related to theorogenic collapse phase of the Araçuaí orogen.
The G5 Supersuite occurs as small or large plutons widespread inthe study area (Fig. 2), and consists of non-foliated monzogranite,syenogranite, and charnockites. Locally, these granitoids are foliatedclose to their contact with the older units. U–Pb data on zircon grainsfrom the study area combined with ages from different parts of theorogen constrain the crystallization ages of the I-type G5 granitoids be-tween 530Ma and 480Ma (e.g., De Campos et al., 2004; Pedrosa-Soareset al., 2011).
3. Analytical methods
Representative samples of the G1 Supersuite granitoidswere chosenfor lithochemical (8 samples), geochronological (U–Pb — 10 samples,Lu–Hf— 4 samples) and isotopic (Sm–Nd, Rb–Sr— 6 samples) analyses.Additionally, two samples of the G2 rocks and one sample of the G5Supersuite were dated (U–Pb zircon analyses). Care was taken to selectfresh and homogenous parts of the samples, avoiding weathered, meta-morphosed, or late veins. All the samples were crushed and milled atthe Departamento de Geologia of the Universidade Federal de OuroPreto, Brazil. Major, trace and rare earth element (REE) concentrationswere determined at the ACME Analytical Laboratory LTDA, Vancouver,Canada. Major oxides were analyzed via inductively coupled plasmaatomic emission spectroscopy (ICP-AES) after fusion with lithiummetaborate–tetraborate and digestion with diluted nitric acid. Traceand REE concentrations were analyzed via inductively coupled plasmamass spectrometry (ICP-MS) adopting the same procedure previouslymentioned as for major oxides. Detection limits are 0.01% for oxidesand 0.1–0.01 ppm for trace, and rare earth elements. The loss on ignition(LOI) was determined by the weighting difference after ignition at1000 °C. The geochemical dataset of samples from the literature as
Table1
Summaryof
theke
ymainfeatures
oftheJequ
itinho
nhapa
ragn
eisses
andG1gran
itic
rocksex
posedin
thestud
yarea
.
Unit
Rock
type
Mineralog
yGeo
chem
istry
Fieldob
servations
Referenc
es
Mainminerals
Accessory
minerals
Jequ
itinho
nhaCo
mplex
Paragn
eiss
Qua
rtz,feldsp
ar,b
iotite,
cordierite,g
arne
t(alm
andine
)Silliman
ite,
grap
hite,o
paqu
esPe
raluminou
sgn
eisses,w
ithpe
litic
grey
wacke
sas
prob
able
protolith
Band
ed,m
igmatized
Gon
çalves-D
iaset
al.
(201
1)Biotitepa
ragn
eiss,A
l-rich
gneiss
Qua
rtz,plag
ioclase,
K-felds
par,
biotite,
cordierite,g
arne
tSilliman
ite,zircon,
apatite
,graph
ite,m
uscovite,
opaq
ues
Foliated,
band
ed,m
igmatized
Paes
etal.(20
10)
Ranc
hoAlegrepluton
sTo
nalite,
minor
gran
odiorite
andmon
zogran
ite
Plag
ioclase,
quartz,K
-felds
par,
biotite
Apa
tite,zirco
n,mon
azite,
garn
et,o
paqu
esSlightly
peraluminou
s,mostly
med
ium-K
calc-alkaline
Isotropicto
folia
ted,
locally
migmatized
Paes
etal.(20
10)
Feliz
burgopluton
sGrano
diorite,
minor
tona
lite
andmon
zogran
ite
Plag
ioclase,
quartz,o
rtho
clase,
microcline,
biotite
Apa
tite,zirco
n,op
aque
s(e.g.,mag
netite),
mon
azite,
titanite,a
llanite
Slightly
peraluminou
s,mostly
med
ium-to
high
-Kcalc-alkaline
Folia
ted,
migmatized
Pedrado
Sino
pluton
sMon
zogran
ite,
gran
odiorite
andtona
lite
Qua
rtz,microcline,
orthoc
lase,
plag
ioclase,
biotite
Garne
t,allanite,e
pido
te,a
patite,zirco
n,op
aque
s,titanite,m
onazite,
mus
covite,
silliman
ite
Not
available
Isotropicto
strong
lyfolia
ted
Carvalho
andPe
reira
(200
0)
SãoVitor
pluton
sGrano
dioritean
dtona
lite,
minor
mon
zogran
itean
dsyen
ogranite
Plag
ioclase,
microcline,
orthoc
lase,q
uartz,biotite
Garne
t,allanite,titan
ite,
apatite,
zircon
,mon
azite,
rutile,o
paqu
esMetalum
inou
sto
slightly
peraluminou
sIsotropicto
folia
ted,
and
band
ed
Topá
ziopluton
sGrano
diorite,
minor
mon
zogran
ite
G1Su
persuite
(Ran
choAlegre,
Felizbu
rgo,Pe
drado
Sino
,São
Vito
r,To
pázio,Bo
mJesusda
Vitó
riapluton
s)
Grano
dioritean
dtona
lite,
minor
mon
zogran
ite
Plag
ioclase,
quartz,o
rtho
clase,
microcline,
biotite,
horn
blen
deZircon
,apa
tite,m
onazite,
garn
et,titan
ite,
allanite,e
pido
te,rutile
,mus
covite,h
ematite,
ilmen
ite,
mag
netite,p
yrrh
otite,
chalco
pyrite
Magne
sian
,mostly
slightly
peraluminou
s,an
dcalcicto
calc-alkalic,
med
ium-to
high
-Kacid-silicicpluton
icrocks
Isotropicto
folia
ted,
locally
migmatized
Thisstud
y
443L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
well as data from this study are presented as Supplementary data —item 1.
U–Pb zircon laser ablation multicollector inductively coupled plasmamass spectrometry (LA–MC–ICP–MS, Neptune Thermo Scientific) analy-seswere carried out at theGeochronology Laboratory of the Universidadede São Paulo, Brazil. Sampleswere prepared for analysis in laboratories ofUniversidade Federal deOuro Preto andUniversidade de São Paulo, Brazil.Zircon grains were separated using conventional methods (crushing,grinding, gravimetric and magnetic-Frantz isodynamic separator), andhandpicked under binocular microscope. Only zircon crystals from theleast magnetic fractions were selected for geochronological analysis. Se-lected zircon grains were mounted in an epoxy disk and polished to ex-pose the interiors. The morphology and internal structure of zirconswere characterized by optical microscopy and cathodoluminescence(CL) imaging. The mount disks were then cleaned and the U–Pb isotopiccompositions of zircon were analyzed. A spot size of 25 μm was appliedto all analyses. Zircon GJ-1 (Jackson et al., 2004) was used as an externalstandard and was analyzed twice every 6 analyses. Analytical spotswere conducted avoiding grain areas with inclusions, fractures andmetamictic structures. Data reduction used the Excel sheet developedby Chemale et al. (2012). For all samples the data of each spot was evalu-ated taking into account the commonPb contents, errors of isotopic ratios,percentages of discordance and Th/U ratios. Selected spots used for agecalculations were those with discordance lower than 10%. Concordia dia-grams were obtained with the software Isoplot/Ex (Ludwig, 2003). U–Pbanalytical data and concordia diagrams are shown as Supplementarydata — item 2 and in Fig. 10.
The Sm–Nd (ID-TIMS) and Sr (natural) isotope analyses were per-formed in the Geochronology and Radiogenic Isotope Laboratory(LAGIR) at UERJ, the Rio de Janeiro State University. The analyzed sam-pleswere 4 granodiorites, 1 tonalite and1 enderbite belonging to theG1Supersuite, and the results are shown as Supplementary Data— item 3and in Fig. 11. Clean rooms were used with positive air pressure anddouble-HEPA air filtering. Each rock powder sample weighting up to50 mg was mixed with an isotope tracer solution of 149Sm and 150Ndin Savillex™ PTFE beakers. Chemical digestion of samples (plus tracer)by a mixture of HF (6 mL) and HNO3 6 N (0.5 mL) on a hot plate, lasted2 periods of 5 days. A primary phase of chromatographic extraction ofSr and REE in HCl used an ion exchange column filled with the BIORADAG50W-X8 resin (100–200 mesh). The extraction of Sm and Nd fromthe REE solution was done with the Eichrom LN-spec resin (150 mesh)in a smaller column. The samples were then loaded separately onto pre-viously degassed Re filament in double assembly, using H3PO4 (1 N) asthe ionization activator. The Sm, Nd and Sr isotope ratios were measuredwith the multi-collector TRITON thermal ionization mass spectrometer(TIMS), in cup configurations of up to 8 Faraday collectors in staticmode. The measured isotope ratios were normalized to the respectiveconstant isotope ratios of 147Sm/152Sm = 0.56083, 146Nd/144Nd =0.7219 and 86Sr/88Sr=0.1194. Correctionswere applied for instrumentalbias, tracer content and blanks below 200 pg Nd, and 70 pg Sm. Repeatedanalyses (n = 140) of the NBS-987 (NIST) and (n = 214) of the Jnd1(Tanaka et al., 2000) standard reference materials yielded mean ratios87Sr/86Sr = 0.710239 ± 0.000007 and 143Nd/144Nd = 0.512100 ±0.000006 (2σ absolute standard errors), respectively. The Sm–Nd TDMmodel ages and the εNd(i) values were calculated using DePaolo's (1981)parameters.
Lu–Hf analyses in zircon were obtained via laser ablation multi-collector inductively coupled plasma mass spectrometry (LA–MC–ICP–MS, Photonmachines 193/Neptune Thermo Scientific) at the IsotopeGeochemistry Laboratory of the Departamento de Geologia of theUniversidade Federal de Ouro Preto, Brazil. The analyzed sampleswere 3 granodiorites and 1 enderbite, and the results are presented asSupplementary data— item 4, and in Fig. 12. Datawere collected in stat-ic mode during 60 s of ablation with a spot size of 50 μm. Nitrogen(~0.080 L/min) was introduced into the Ar sample carrier gas. Typicalsignal intensity was ca. 12 V for 180Hf. The isotopes 172Yb, 173Yb and
Fig. 3. Petrographic characteristics of the studied G1 granitoids. A–B) Partially molten features preserved in the G1 rocks (samples LG59 and LG63, respectively). Arrows indicate diffuseneosomes likely formed at the end of regional deformation; C) SEM image of a granitoid samplewithwell preserved euhedral plagioclase grains (sample LG173); D)Microscopic images ofelongated quartz grains showing undulose extinctions (transmitted light in the left and polarized light in the right— sample LG31); E)Mechanical twinning developed in plagioclase crys-tal from a tonalite belonging to the Rancho Alegre pluton (sample LG63). Symbols refer to quartz (Qz), biotite (Bt) and plagioclase (Pl).
444 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
175Luwere simultaneouslymonitored during each analysis step to allowfor correction of isobaric interferences of Lu and Yb isotopes on mass176. The 176Yb and 176Lu were calculated using a 176Yb/173Yb of0.796218 (Chu et al., 2002) and 176Lu/175Lu of 0.02658 (JWG in-housevalue). The correction for instrumental mass bias utilized an exponen-tial law and a 179Hf/177Hf value of 0.7325 (Patchett and Tatsumoto,1980) for correction of Hf isotopic ratios. The mass bias of Yb isotopesgenerally differs slightly from that of theHf isotopeswith a typical offsetof the βHf/βYb of ca. 1.04 to 1.06 when using the 172Yb/173Yb value of1.35274 from Chu et al. (2002). This offset was determined for each an-alytical session by averaging the βHf/βYb of multiple analyses of theJMC 475 solution doped with variable Yb amounts and all laser ablationanalyses (typically n N 50) of TEMORA zircon with a 173Yb signal inten-sity of N60mV. Themass bias behavior of Luwas assumed to follow thatof Yb. The Yb and Lu isotopic ratios were corrected using the βHf of theindividual integration steps (n= 60) of each analysis divided by the av-erage offset factor of the complete analytical session. In the course ofanalysis, secondary standards such as Plešovice, TEMORA, 91500, GJ1,and BB9 yielded 176Hf/177Hf ratios of 0.282477 ± 0.000015 (2σ, n =22), 0.282657 ± 0.000017 (2σ, n = 23), 0.281663 ± 0.000016 (2σ,n = 21), 0.282006 ± 0.000015 (2σ, n = 3), and 0.282298 ± 0.000023(2σ, n = 10), respectively. These ratios are in good agreement withthe recommended values (e.g., Griffin et al., 2006; Wu et al., 2006;Morel et al., 2008; Sláma et al., 2008).
4. Results
4.1. Petrography
The selected samples for detailed petrographic studies were collect-ed in the G1 bodies known as Rancho Alegre, Felizburgo, Pedra do Sino,Bom Jesus da Vitória, São Vítor and Topázio plutons (Fig. 2). The firstfour plutons, exposed in the central-northeastern sector of the studyarea, intrude rocks of the Jequitinhonha Complex, whereas the lasttwo cut Rio Doce Group rocks in the southwestern portion of thestudy area (Fig. 2). These plutons are composed mainly of tonalites,granodiorites and minor monzogranites (Table 1). The dominantlithotypes are light to medium gray, fine to coarse-grained, and consistmostly of plagioclase (33–52%), quartz (23–48%), K-feldspars (10–38%;microcline and orthoclase), biotite (2–25%), and minor hornblende(~3%), the latter mineral being restricted to the Bom Jesus da Vitóriapluton. Zircon, apatite, titanite, garnet, epidote, hematite, magnetiteand ilmenite are common accessory minerals, while rutile, muscovite,allanite, monazite, pyrrhotite, pyrite and chalcopyrite are restricted tofew samples. Muscovite and carbonates are secondary products of pla-gioclase breakdown. Chlorite occurs as replacement of biotite.
The studied granitoids are, in general, strongly foliated, showing lo-callymigmatitic features (Fig. 3A–B). Undeformed phases also occur, es-pecially in the central portion of the plutons. Preserved igneous features
Fig. 4. A) R1–R2 diagram showing the variability of G1 rocks forming the northern segment of the Rio Docemagmatic arc (after De La Roche et al., 1980). Filled circles refer to data of Paeset al. (2010) fromG1 rocks that are exposed in the study area, while plus sign represent the analyses of this study. Ab= albite; An50= plagioclase An50; Or= orthoclase. B) AFM diagramshowing the distribution of the studied G1 rocks,which is similar to a typical calc-alkaline series (after Rickwood, 1989). C) Characteristics of the rocks forming the northern portion of theRioDoce arc based on Sand's index (afterManiar and Picolli, 1989). CAG=continental arc granitoids; CCG=continental collision granitoids (seeManiar and Picolli (1989) and referencestherein for data source). Dashed line marks the transition between I- and S-type granites according to Chappell and White (1974). D) Variation of the Aluminum Saturation Index (A/CNK = mol. Al2O3/CaO + Na2O + K2O) versus maficity (Fe + Mg) for the studied G1 Supersuite granitoids. I/S-type granitoid boundary at ASI = 1.1 from Chappell and White (1974).E) FeOtot/(FeOtot + MgO) versus weight percent SiO2 diagram showing the predominant magnesian character of the rocks forming the studied G1 Rio Doce arc rocks. Continuous linemarks the boundary between ferroan and magnesian plutons of Mesozoic batholiths from North America (see Frost et al., 2001 and references therein for details). F) Plot ofNa2O+K2O–CaO (MALI) versus SiO2 (wt.%) showing the spread distribution of the Rio Doce arc rocks. For comparison Famatinian, Paleoproterozoic,Mesozoic Cordilleran, and 1155 anal-yses of I-type granitoids from Lachlan Fold Belt of southeastern Australia are also shown (after Frost et al., 2001). Continuous line constrain the composition range of 344 samples fromCordilleran Mesozoic batholiths of North America; Short dashed line constrain composition range of 12 samples from the Ordovician Famatinian magmatic arc (see Pankhurst et al.,1998 for data source); and long dashed line constrain the composition range of 41 samples from the Paleoproterozoic magmatic arc (for source of data see Dunphy and Ludden, 1998),Ungava orogen, Canada (see Frost et al., 2001, and references therein for additional information). In diagrams B to F, shaded gray area corresponds to the compositional range of ca.200 granitic samples of the central–southern segments of the Rio Doce arc rocks (for source of data see Gonçalves et al., 2014, and Supplementary data — item 1).
445L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
Fig. 6. Rare earth elements diagram normalized by chondrite, according to values ofNakamura (1974). Shadow area delimits the REE concentration for data from this study.See text for further details.
447L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
such as zoned and tabular plagioclase crystalswere observed in Topázio,São Vítor and Pedra do Sino plutons (Fig. 3C). Solid-state deformationfeatures, such as elongated grains, undulose extinctions, mechanicaltwinning, deformation bands, and occasionally quartz subgrains domi-nate the texture of the rocks (Fig. 3D–E). The associated metamorphicfoliation is marked by the alignment of biotite and hornblende.Brown-reddish biotite is mostly euhedral to subhedral and shows pleo-chroic halos given by metamictic zircon and monazite grains. Horn-blende grains occur as anhedral to subhedral crystals.
4.2. Geochemistry
Eight samples of G1 Supersuite rocks were analyzed for major andtrace elements (Supplementary data — item 1). Additional major andtrace element compositions of 17 samples from the study area analyzedby Paes et al. (2010) are also shown in the diagrams of Fig. 4. Forcomparison, we also plotted approximately 200 chemical analyses of gra-nitic rocks belonging to the central–southern portions of the Rio Doce arc(shaded areas in the diagramsof Fig. 4) (seeGonçalves et al., 2014 for datasource), 50 chemical compositions of samples from Grosse et al. (2011),mainly for major element comparisons, and 12 from Pankhurst et al.(1998), mainly for trace element comparisons, belonging to theFamatinian magmatic arc, which show similar lithochemical attributeswhen compared with the Rio Doce magmatic arc (Fig. 5).
The analyses of this study and those of Paes et al. (2010) show arange of compositions from tonalite to monzogranite. Granodiorite(52%) and tonalite (32%) represent by far the most common rocks(Fig. 4A). They have a SiO2 content ranging from 61.66% to 78.27%(see Supplementary data — item 1 and Paes et al., 2010). The samplecompositions are quite homogenous, defining a calc-alkaline series(Fig. 4B), with a dominant peraluminous affinity (average ASI = 1.07),with the exception of sample LG257 (ASI = 0.94) (Fig. 4C–D). All sam-ples aremagnesian, except sample LG259which is ferroan (Fig. 4E), andrange from calcic to alkali-calcic. Calcic (68%) and calc-alkaline (24%)rocks form the bulk of the G1 bodies in the study area (Fig. 4F).
The samples show a clear negative correlation between TiO2, Al2O3,Fe2O3 and MgO with SiO2, and although more scattered, CaO (1.84–5.81 wt.%) data also show a slightly negative correlation with SiO2
(Fig. 5A–E), suggesting fractionation of mafic minerals. On the otherhand, a slightly positive correlation between SiO2 and Na2O is observed(Fig. 5F). No clear correlation is observed between SiO2 and K2O, al-though K2O generally increaseswith SiO2 (Fig. 5G). The Alumina Satura-tion Index does not show correlation with SiO2 (Fig. 5H).
The analyzed samples display REE patterns characterized by enrich-ment in light rare earth elements (LREE) over heavy rare earth elements(HREE), with (La/Yb)N ratios ranging between ~11 and ~63 (Supple-mentary data — item 1, Fig. 6). All samples record a negative Eu anom-aly, with Eu/Eu* ratios ranging from ~0.33 to ~0.79 (Supplementarydata — item 1). In Primitive Mantle-normalized incompatible elementspidergrams (Fig. 7A–D), the G1 rocks show overall large ion lithophileelements (LILE)-enriched patterns (Rb, Th), with values between 100and 600 times higher than the Primitive Mantle values (Sun andMcDonough, 1989). This pattern combined with positive Th, La, Ce, Pand Zr anomalies suggests important crustal sourcing. Other commonchemical characteristic of the analyzed rocks is the negative Nb, Ta, Srand Ti anomalies (Fig. 7B), which is normally referred to as an “arc-like signature” (see Niu et al., 2013, and references therein for details).
4.3. Geochronology
Age determinations were performed on 10 rock samples (Figs. 2, 8–10): 7 samples from the G1 Supersuite; 2 samples from the G2
Fig. 5. A–H) Harkermajor oxides variation diagrams for the studied G1 rocks and Famatinian arare from Rickwood (1989). I/S-type granitoid boundary at ASI = 1.1 from Chappell andWhitebiotite; Hbl = hornblende; Ms = muscovite; Ttn = titanite; Aln = aluminosilicates; Ep = epi
Supersuite; and 1 sample from the G5 Supersuite. The analyzed samplescorrespond to one enderbite, one tonalite, seven granodiorites, and onesyenogranite. A summary of the main features of the dated samples areshown in Tables 2 and 3 and Figs. 8 to 10.
4.3.1. G1 Supersuite (samples LG21, LG27, LG28, LG31, LG49, LG63, andLG173)
The studied G1 Supersuite plutons showed that six samples (LG21,LG27, LG28, LG49, LG63 and LG173) have zircon grains with inheritedcores on CL images (Fig. 9). Attempts to date these cores failed and nocoherent age data could be obtained, however apparent ages (from958 Ma to 2574 Ma) are suggestive of inheritance from rocks of the Je-quitinhonha Complex (e.g., Gonçalves-Dias et al., 2011). Three samples(LG21, LG31 and LG173) gave ages of 574 ± 7 Ma, 575 ± 6 Ma, and584± 13Ma, respectively, which are interpreted as their crystallizationages (Fig. 10A, D, G and Tables 2 and 3). On the other hand, four samples(LG27, LG28, LG49 and LG63) have zircon grains, whichwere divided intwo populations, giving ages of ca. 618–595 Ma for crystallization(Fig. 10B-I, C-I, E-I, F-I and Tables 2 and 3) and ca. 555–589Ma for over-growths or new crystallized zircon grains (Fig. 10B-II, C-II, E-II, F-II andTables 2 and 3) interpreted as a migmatization event affecting therocks. Assuming that this interpretation is correct, the obtained resultsmore likely reflect the regional metamorphic event that took place inthe Araçuaí orogen between 590 Ma to 545 Ma, peaking at 575 Ma(Pedrosa-Soares et al., 2001, 2011; Silva et al., 2011; Gradim et al.,2014; Peixoto et al., 2015).
4.3.2. G2 Supersuite (samples LG12 and LG161)The zircon grains extracted from sample LG12 (Caraí pluton) com-
prise mainly short-prismatic, and minor elongated grains (up to422 μm) with length/width ratios in average of 4:1, although higher ra-tios up to 6:1 can be observed. The axes of zircon grains range from37 μm to 422 μm.Most grains are dark-gray and some displaymagmaticoscillatory zoningwith inherited cores on CL images (Fig. 9). Several at-tempts to date these cores were unsuccessful due to the high level ofdiscordance, although apparent ages ranging mainly from ca. 800 Mato 1200 Ma could be observed. Most zircon grains have Th/U ratios
c granitoids from Sierra de Velasco (Grosse et al., 2011). Fields in the K2O vs. SiO2 diagram(1974). IG= I-type granitoids; TG= transitional granitoids; SG= S-type granitoids. Bt =dote; Crd = cordierite.
Fig. 7. Multi-element spidergrams normalized to the Primitive Mantle (normalizing values after McDonough and Sun, 1995) of the (A) central–southern G1 Supersuite granites (seeGonçalves et al., 2014 for data source); (B) Studied northern G1 Supersuite rocks. Shadow area delimits the granitoid compositions studied by Paes et al. (2010); (C) Granitoid rocks ofthe Paleoproterozoicmagmatic arc from theUngava orogen (seeDunphy and Ludden, 1998 for data source), and (D) Granitoid compositions from the Famatinianmagmatic arc (see Pank-hurst et al., 1998 for data source).
448 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
ranging from 0.10 to 0.83 (Supplementary data— item 2). Thirteen con-cordant to near-concordant (b4% disc.) spot analyses from 12 zircongrains yielded a concordia age of 538 ± 8 Ma (MSWD = 0.21)(Fig. 10H). This age is interpreted as the best approximation of the crys-tallization age of this sample.
Sample LG161 (Wolff pluton) contains short- and long-prismaticzircon grains, with length/width ratios in average of 4:1, althoughhigher ratios up to 6:1 are observed. The axes of zircon grains rangefrom47 μmto 330 μm.They are light-gray and somegrains displaymag-matic oscillatory zoning with inherited cores on CL images (Fig. 9). At-tempts to obtain ages from these cores failed due to the high level ofdiscordance, although apparent ages mainly between ~960 Ma and~2000 Ma are found. Four zircon grains have Th/U ratios between 0.38and 1.22, while two other grains have Th/U ratios of 0.02 and 0.03 (Sup-plementary data — item 2). Six concordant to near-concordant spotanalyses obtained in the same number of zircon grains yielded aconcordia age of 563 ± 9 Ma (MSWD = 0.54) (Fig. 10I). We interpretthis age as the magmatic crystallization age of this sample.
4.3.3. G5 Supersuite (sample LG29)The zircon grains extracted from sample LG29 (Fazenda Liberdade
pluton) consist of predominantly short-prismatic grains, with length/width ratios in average of 2.6:1, although higher ratios up to 5:1 are ob-served. The axes of zircon grains range from 45 μm to 245 μm. Mostgrains are light-gray, and seldom display magmatic oscillatory zoningwith no inherited cores on CL images (Fig. 9). Many zircon grains haveTh/U ratios higher than 1 (Supplementary data — item 2). Thirty one
concordant (b2% disc.; except one grainwith 4% disc.) spot analyses ob-tained in the same number of zircon grains yielded a concordia age of526 ± 5 Ma (MSWD = 0.12) (Fig. 10J). This age is interpreted as thebest approximation of the crystallization age of this sample.
4.4. Rb–Sr and Sm–Nd isotope data
Six whole-rock samples of the G1 Supersuite granitoids were ana-lyzed for Rb–Sr and Sm–Nd isotopes (Supplementary data — item 3).They comprise 4 granodiorites (samples LG21, LG28, LG49 andLG173), 1 tonalite (sample LG63), and 1 enderbite (sample LG31).
The 87Sr/86Sr isotopic ratios, when calculated to their crystallizationage (ca. 630–580 Ma), range from 0.7059 to 0.7121 (Fig. 11). The87Rb/86Sr ratios vary between 0.6979 and 0.9697 (Supplementarydata — item 3). The high 87Sr/86Sr ratios and the lack of correlationwith the 87Rb/86Sr ratios suggest involvement of metasedimentary ma-terial in the source of the G1 rocks by variable degrees of melting-assimilation.
The measured 143Nd/144Nd ratios are quite homogeneous and showa range of data between 0.511904 and 0.511983. Similarly, the mea-sured 147Sm/144Nd ratios are uniform and have values ranging between0.081 and 0.117 (Supplementary data— item3). The εNd(i) values, calcu-lated for the crystallization age of the studied rocks, vary between−5.7and −7.8 (Supplementary data — item 3; Fig. 11). The distribution ofTDM model ages spans mainly from 1.36 Ga to 1.68 Ga, showing thatStatherian to Calymmian rocks of the Jequitinhonha Complex might
Fig. 8.Macroscopic photographs of thedated samples: A–G) rock samples belonging to theG1Supersuite;H–I) samples from theG2 Supersuite and J) granitoid sample belonging to theG5Supersuite (see Fig. 2 for sample location and text for further details).
449L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
450 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
be involved somehow in the formation of the northern segment of theRio Doce magmatic arc.
4.5. Lu–Hf isotope data
Four representative samples of the G1 Supersuitewere chosen for Hfisotope analyses (see Fig. 2 for sample locality). Taken together, the an-alyzed zircon grains (9 spots in the sample LG21, 16 in the sample LG27,22 in the sample LG31 and 22 spots in the sample LG49) have quite ho-mogeneous Hf isotopic compositions, with 176Hf/177Hf ratios rangingfrom 0.282083 to 0.282278 and εHf(t) values from −5.2 to−11.7 (av-erage of −7.1) (Supplementary data — item 4, and Fig. 12).
Zircon grains from the granodiorite belonging to the Topázio pluton(sample LG21, Fig. 2) have 176Hf/177Hf ratios of 0.282239–0.282278 andεHf(t) values from−5.2 to−6.5, with average of−6.0 (Supplementarydata — item 4). The two distinct groups of previously described zircongrains (Section 4.3.1) from the Pedra do Sino Granodiorite (sampleLG27, Fig. 2, Table 3) show quite homogeneous Hf isotopic composi-tions, with 176Hf/177Hf ratios ranging from 0.282144 to 0.282277 andεHf(t) values from−5.5 to−10.2, and average of−7.4 (Supplementarydata — item 4). Zircon grains from the enderbite belonging to the BomJesus da Vitória pluton (sample LG31, Fig. 2, Table 3) have 176Hf/177Hfratios ranging between 0.282158 and 0.282246, with εHf(t) valuesfrom −6.1 to−9.3, and average of −7.7 (Supplementary data — item4). Similarly to the sample of the Pedra do Sino Pluton (LG27), thetwo zircon populations from the Rancho Alegre pluton (Fig. 2, Table 3)granodiorite (sample LG49) exhibit quite homogeneous Hf isotopic
Fig. 9. Cathodoluminescence images showing the external characteristics and internal structurLG63, and LG173); G2 Supersuite (samples LG12 and LG161), and G5 Supersuite (sample LG29
compositions, with no clear difference among the various zircon grains.Zircon grains from this sample yielded 176Hf/177Hf ratios from 0.282083to 0.282258, with εHf(t) values ranging from−5.4 to−11.7, with aver-age of −7.2 (Supplementary data — item 4, and Fig. 12).
5. Discussion
The rocks of the G1 Supersuite that represent the termination of theRio Doce magmatic arc in the northern Araçuaí orogen seem to charac-terize a special assemblage of granitoids. Our results indicate that theircrystallization ages range from ca. 630 Ma to 570 Ma (within errors),predating (with some overlap) the ages of typical collisional,peraluminous (ASI N 1.1) granites of the G2 Supersuite, dated between590 Ma and 545 Ma (e.g., Pedrosa-Soares et al., 2011; Gradim et al.,2014). The obtained ages from this study are in good agreement withprevious ages determined for the Rio Doce arc (e.g., Pedrosa-Soareset al., 2011; Gonçalves et al., 2014 and references therein). The G2Supersuite rocks share similarities with typical collisional granitoids.Consisting mainly of S-type peraluminous two-mica granites and bio-tite–garnet–cordierite–sillimanite leucogranites, exhibiting chemicalpatterns of calc-alkaline to alkali-calcic and high-K granitoids, the G2Supersuite greatly differs from the studied G1 rocks (see Gradim et al.,2014 for a synthesis review on the collisional granitoids from theAraçuaí orogen).
The U–Pb ages obtained in this study for the G1 Supersuite charac-terizes two rock groups: i) older granitoids crystallized between 590and 630 Ma and migmatized at ca. 585 Ma; ii) younger granitoids
es of zircon grains extracted from: G1 Supersuite (samples LG21, LG27, LG28, LG31, LG49,). Spot sizes: 25 μm (U–Pb analyses) and 50 μm (Lu–Hf analyses).
Fig. 10. U–Pb concordia diagrams of LA–MC–ICP–MS analyses of zircon from G1 Supersuite: A–G) samples LG21, LG27, LG28, LG31, LG49, LG63, and LG173; G2 Supersuite: samples LG12 and LG161 (H–I), and G5 Supersuite: sample LG29 (J).
451L.G
onçalvesetal./G
ondwana
Research36
(2016)439–458
452 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
crystallized between 570 and 590Ma that do not show any evidence formigmatization. The first group of granitoids, exposed in the northern-most extremity of the arc and intruding the Jequitinhonha Complexrocks (Fig. 2), includes the Pedra do Sino and Rancho Alegre plutons(samples LG27, LG28, LG49, and LG63). The close association with the
water-rich Jequitinhonha rocks could be responsible for triggering themigmatization of the granitoids in the area.
The second group is exposed in the southwestern (sample LG21) andnortheastern (samples LG31 and LG173) portions of the study area andincludes the Topázio, Pedra do Sino and Bom Jesus da Vitória plutons.
Table2
Summaryof
themainfeatures
oftheda
tedsamples.
Pluton
(sam
ple)
UTM
location
Rock
Mineralog
yFieldob
servations
Age
anderror(M
a)(M
SWD)
Mainminerals
Accessory-secon
dary
minerals
Mag
matic
Migmatitic
G1 Su
persuite
Topá
zio(LG21
)24
6496
/804
6722
Grano
diorite
Plagioclase,qu
artz,K
-feldspa
r(m
ostly
microcline),and
biotite
Zircon
,apa
tite,h
ematite,
ilmen
ite
Grayish
,med
ium-to
coarse-grained
;foliatedto
band
ed57
4±
7(0
.052
)–
Pedrado
Sino
(LG27
)29
4499
/810
2392
Grano
diorite
Plag
ioclase,
quartz,K
-felds
par,
andbiotite
Zircon
,mon
azite,
hematite,
minor
pyrrho
tite,cha
lcop
yrite,
mus
covite
and
carbon
ates
Grayish
,fine
-to
coarse-grained
;mm-to
cm-sized
oriented
K-felds
pars
areindicative
ofmag
matic
flow;m
igmatized
616±
12(0
.077
)55
5±
7(0
.003
7)
Ranc
hoAlegre(LG28
)29
5463
/813
1093
Grano
diorite
Plag
ioclase,
quartz,K
-felds
par,
andbiotite
Garne
t,zircon
andmus
covite
Grayish
,mostlymed
ium-grained
;foliatedan
dmigmatized
595±
13(0
.007
1)57
6±
7(0
.033
)Bo
mJesu
sda
Vitória
(LG31
)32
7321
/813
8883
Ende
rbite
Ortho
pyroxe
ne,b
iotite,
horn
blen
de,p
lagioc
lase,a
ndqu
artz
Zircon
,epido
te,a
llanite,apa
tite,and
minor
hematite,
ilmen
ite,
chalco
pyrite,
pyrrho
tite
andpy
rite
Green
ishfolia
ted;
fineto
med
ium-grained
575±
6(0
.17)
–
Ranc
hoAlegre(LG49
)30
6086
/814
5456
Grano
diorite
Plag
ioclase,
quartz,K
-felds
par,
andbiotite
Garne
t,titanite,a
patite,zirco
nan
dmus
covite
Grayish
,mostlymed
ium-grained
;foliatedan
dmigmatized
618±
9(0
.000
38)
589±
7(0
.26)
Ranc
hoAlegre(LG63
)29
6620
/813
8129
Tona
lite
Plagioclase,qu
artz,K
-feldspa
r(m
ostly
orthoclase),an
dbiotite
Apa
tite,titan
ite,
zircon
andmus
covite
Grayish
,fine
-to
coarse-grained
;foliatedan
dmigmatized
607±
9(0
.094
)58
7±
10(0
.65)
Pedrado
Sino
(LG17
3)31
8559
/811
1399
Grano
diorite
Plagioclase,qu
artz,K
-feldspa
r(m
ostly
microcline),and
biotite
Apa
tite,zirco
n,mag
netite,h
ematite,
carbon
ates
andmus
covite
Grayish
,med
ium-to
coarse-grained
;isotrop
icto
strong
lyfolia
tedwithde
velopm
entof
shea
rzo
nes
584±
13(0
.55)
–
G2 Su
persuite
Caraí(LG
12)
1936
12/802
6021
Syen
ogranite
Plag
ioclase,
microcline,
quartz
andbiotite
Cordierite,g
arne
t,zircon
,hem
atite,
mus
covite,carbo
natesan
dch
lorite
Grayish
,med
ium-to
coarse-grained
;slig
htly
folia
tedto
band
ed53
8±
8(0
.21)
539±
11(0
.24)
Wolff(LG16
1)27
8324
/809
1036
Grano
diorite
Plagioclase,qu
artz,K
-feldspa
r(m
ostly
orthoclase),an
dbiotite
Apa
tite,g
arne
t,zircon
,hem
atite,
ilmen
ite,
cordierite,sillim
anite,
carbon
ates
and
mus
covite
Grayish
,med
ium-to
coarse-grained
;slig
htly
folia
ted
563±
9(0
.54)
–
G5 Su
persuite
Fazend
aLibe
rdad
e(LG29
)29
4908
/813
6533
Grano
diorite
Plag
ioclase,
quartz,K
-felds
par
(mostlymicrocline),a
ndbiotite
Apa
tite,titan
ite,
epidote,
rutile,zirco
n,mag
netite,ilm
enite,
carbon
ates
and
mus
covite
Grayish
,med
ium-to
coarse-grained
;isotrop
icwithroun
ded
cm-to
decimetric-size
dmafi
cen
clav
es;felds
parcrystals
markamag
matic
flow
526±
5(0
.12)
–
453L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
The Topázio pluton intrudes the metasedimentary rocks of the Rio DoceGroup (Fig. 2). Sample LG21was collected in the central part of the plutonand do not show evidence for migmatization. The age of 574 ± 7 Mayielded by this sample is among the youngest ages obtained for the G1magmatism in the Araçuaí orogen (e.g., Pedrosa-Soares et al., 2011;Gonçalves et al., 2014, and references therein). In addition, this agematches quite well, within errors, with the crystallization ages of plutonsexposed nearby: the São Vítor (585 ± 7 Ma, Mondou et al., 2012),Brasilândia (581 ± 11 Ma, Tedeschi, 2013), and Guarataia plutons(576 ± 9 Ma, Tedeschi, 2013) also ascribed to the G1 Supersuite(e.g., Pedrosa-Soares et al., 2011; Gonçalves et al., 2014).
Sample LG173 (ca. 585 Ma) from the Pedra do Sino pluton exposedin the northeastern sector of the studied area (Fig. 2) does not show ev-idence for migmatization (Table 2). The absence for migmatizationcould be due to lower contents of fertile minerals when compared tosample LG27, extracted in the same region (Fig. 2). This also occurswith sample LG31 (ca. 575 Ma). In this case, the absence ofmigmatization could be caused by the enderbitic composition of theBom Jesus da Vitória pluton, requiring higher P–T conditions for the par-tial melting process.Worth noting, the crystallization age of this sampleis similar to other crystallization ages already obtained for Opx-bearingrocks belonging to the Rio Doce magmatic arc (Mondou et al., 2012;Tedeschi, 2013; Gonçalves et al., 2014).
Chemically, the investigated G1 Supersuite rocks correspond to amagnesian, slightly peraluminous, calcic- to calc-alkaline, medium- tohigh-K acid–silicic series (Figs. 4A–F, and 5G–H). The εNd values and ini-tial 87Sr/86Sr ratios are intermediate between typical I- and S-type gran-ites (Fig. 11). The Hf isotope data indicates a long period of crustalextraction with TDM ages of ca. 1.5–1.6 Ga, and εHf values ranging from−5.2 to −11.7, with an average of−7.1 (Fig. 12). Importantly, typicalI- and S-type granites are characterized by a large variability in the Hfisotopic system, and show some overlap in their εHf values (see Kempet al., 2007; Villaros et al., 2012 for a comprehensive review). Thus,the Hf isotope results obtained in this study do not bring additional in-formation for classifying the northernmost G1 rocks either as I- or S-type granitoids.
The studied G1 rocks show features that are particular to arc-relatedgranites, which are:
1) They are calcic (68%) and calc-alkaline (24%), with minor alkali-calcic (8%) members, occasionally containing titanite or garnet, butcommonly without hornblende (Table 2). Granitoids that containhornblende (~3%) are similar to G1 granitoids exposed south ofthe study area (see Gonçalves et al., 2014 for details), and are alsochemically similar to I-type granites (e.g., Frost et al., 2001;Chappell et al., 2012; among others).
2) Most granitoids are slightly peraluminous (average ASI = 1.07)(Figs. 4C–D and 5). Their REE patterns are marked by relative LREEenrichment and HREE depletion, producing a concave-up pattern(Fig. 6). They all have moderate negative Eu and Sr anomalies, likelysuggesting feldspar fractionation at the source. Significantly, all gra-nitic rocks sampled here display very similar mantle normalized el-emental patterns (Fig. 7).
3) In diagrams of SiO2 content against Fe number (Fe#) (Fig. 4E) andNa2O + K2O–CaO (Fig. 4F), the granitic rocks plot close to thoseemplaced in subduction-related environments (i.e., Cordilleran-type granites; see Frost et al., 2001 for more details).
The northernmost G1 rocks show similarities and differences withrocks of the remaining segments of the Rio Doce arc. South of thestudy area, the G1 Supersuite is composed of large plutons and batho-liths of medium- to high-K, calc-alkaline, tonalitic to granitic composi-tions, and their Opx-bearing equivalents, generally containingabundant mafic and dioritic facies and enclaves (Nalini et al., 2005;Pedrosa-Soares et al., 2011; Gonçalves et al., 2010, 2014). A few maficto intermediate plutons rich in gabbroic to enderbitic facies are alsopresent (Novo et al., 2010; Tedeschi, 2013; Gonçalves et al., 2014).
Table 3Summary of the main characteristics of zircon grains from the studied G1 granitic rocks.
Pluton(sample)
Rock Zircon information
Morphology Size Internal features Length/widthratios
Th/U ratios Observations
Topázio (LG21) Granodiorite Short and longprismatic grains
Up to 430 μm Medium-gray with few grainsdisplaying magmatic oscillatory zoningwith inherited cores on CL images
4:1 to 6:1 0.20 to 1.32, with fewexceptions close to 0.05
No metamorphicovergrowth
Pedra do Sino(LG27)
Granodiorite Short-prismaticand roundedgrains
55 μm–320 μm Light to medium-gray with few grainsexhibiting magmatic, sometimeschaotic, oscillatory zoning on CL images
3:1 to 4:1 0.20 to 1.76 (firstpopulation) 0.15 to 1.19(second population)
Inherited cores withapparent age of ca.1070 Ma
Rancho Alegre(LG28)
Granodiorite Short prismaticgrains
50 μm–280 μm, inaverage b 200 μm
Light to medium-gray with some grainsdisplaying magmatic oscillatory zoningand inherited cores on CL images
2.5:1 to 5:1 0.10 to 0.35 (firstpopulation) 0.04 to 0.93(second population)
Inherited cores withapparent ages from928 Ma to 2553 Ma
Bom Jesus daVitória (LG31)
Enderbite Short and longprismatic grains
47 μm–460 μm Light-gray with most grains displayingmagmatic oscillatory zoning with noinherited cores on CL images
3:1 to 6.5:1 0.23 to 1.15, with oneexception of 0.12
Sometimes roundedboundaries, howeverno metamorphicovergrowth
Rancho Alegre(LG49)
Granodiorite Short-,long-prismaticand largeelongated grains
Up to 345 μm Medium-gray with some grainsdisplaying magmatic oscillatoryzoning and few grains with inheritedcores
4:1 to 6:1 0.10 to 0.32 (firstpopulation) 0.11 to 1.76(second population)
Inherited cores withapparent ages of958 Ma, 1560 Maand 2405 Ma
Rancho Alegre(LG63)
Tonalite Long andelongated grains
28 μm–436 μm Medium-gray with some grainsdisplaying magmatic oscillatoryzoning on CL images; few grains haveinherited cores and exhibit corrodedand rounded boundaries
4:1 to 9:1 0.10 to 1.07 (firstpopulation) 0.17 to 0.80(second population)
Inherited cores withapparent ages of1283 Ma and 2574Ma; no metamorphicovergrowth
Pedra do Sino(LG173)
Granodiorite Short and longprismatic grains
60 μm–310 μm Medium-gray, show slightlymagmatic oscillatory zoning andsome grains with inherited cores onCL images
4:1 0.11 to 1.83 Inherited cores withapparent ages of1091 Ma, 1256 Maand 2255 Ma
454 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
Particularly, they are metaluminous to slightly peraluminous (averageASI = 1.02; obtained from 25 samples studied by Gonçalves et al.,2014), and hornblende–biotite bearing granitoids, while the northern-most G1 rocks are essentially slightly peraluminous (average ASI =1.07), and mostly biotite-bearing granitoids. The zircon U–Pb
Fig. 11. Epsilon Nd versus Sr/Sr diagram for granitoids from the Rio Doce magmatic arc.DM stands for depletedmantle and CC for the continental crust. The hyperbola representsa putativemixing line between the average Ordovician turbidite (av. sEd.) and the deplet-edmantle end-member (DM) determined byMcCulloch and Chappell (1982), and encap-sulates most hornblende and cordierite granites (black crosses represent 10% mixingintervals). Isotopic compositions for granites from the Lachlan fold belt as well as datafrom G1 granitoids that are exposed south of the studied area are plotted for comparisons(modified from Kemp and Hawkesworth, 2003).
crystallization ages of the previously studied G1 Supersuite rocksrange predominantly from 630 Ma to 580 Ma, similar to the studiedgranitoids, but contain large amounts of ca. 2.1 Ga inherited zircongrains that are probably derived from the basement units (Noce et al.,2007; Pedrosa-Soares et al., 2011; Gonçalves et al., 2014). The ca.2.1 Ga inherited zircon ages were not found in the northernmost G1rocks. The εNd values of the previously studied G1 Supersuite rocksrange from −6 to −13. Their initial 87Sr/86Sr ratios vary from 0.7088to 0.7233. The Sm–Nd isotope data indicates a long period of crustal ex-traction with TDM ages of ca. 1.39–2.26 Ga (Nalini, 1997; De Camposet al., 2004;Martins et al., 2004; Tedeschi, 2013 and references therein).Therefore, the Nd and Sr isotopic results obtained in this study are ingood agreement with data from the literature on the Rio Doce arcrocks (Supplementary data — item 3, and Fig. 11).
The petrographic, geochemical and isotopic characteristics of thegranitic rocks forming the northern end of the Rio Doce arc clearly indi-cate that they share a series of features with both I-type and S-typegranites (cf. Chappell and White, 2001). Furthermore, the lack of horn-blende (with the exception of sample LG31, Table 2), the local occur-rence of titanite, the peraluminosity affinity and scarcity of muscovite,along with the occurrence of both meta-igneous and metasedimentaryenclaves indicate that they are inmany aspects similar to the transition-al I/S-type granitoids described byGrosse et al. (2011) in the Famatinianmagmatic arc of northwesternArgentina. In reality, the studiedG1 rocksshow features that are intermediate between I-type and the transitionalI/S-type granites from the Famatinian magmatic arc (see Grosse et al.2011).
Using of the A–B diagram of Debon and Le Fort (1983) and Villasecaet al. (1998), the studied G1 rocks plot in distinct fields. The G1 granitesare in general moderate to low peraluminous rocks (88% of the sam-ples), and lie between typical I and the I/S-type granites from theFamatinian arc (Fig. 13). In addition, regardless of their felsic or maficcharacter the G1 granitoids show fairly constant peraluminosity. In-deed, the composition plots of the studied rocks on the A–B diagramgenerate a trend comparable to experimental melting of greywackesand amphibolites (Fig. 13). This fact suggests that a progressively incor-poration of metasedimentary material in the melts result in
Fig. 12. Hf isotopic composition of zircon grains from the studied granitoids plotted against their ages. DM= depleted mantle; CHUR= chondritic uniform reservoir.
455L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
compositional variations between pure I-type rocks and transitional I/Sgranites. This interpretation is reinforced by negative εHf and εNd values,ranging between −5.2 and −11.7, TDM ages of ca. 1.5–1.6 Ga, lack ofhornblende and strong crustal initial 87Sr/86Sr ratios obtained for thestudied G1 Supersuite rocks (Supplementary data — items 3 and 4,and Figs. 11, 12, and 13).
Additionally, as shown by the binary diagrams of Fig. 14A–D, thestudied granitoids plot almost entirely in the field of dehydration-melting of amphibolites. Only a few samples plot in the fields ofdehydration-melting of greywackes and none of them in the felsic ormafic pelitefields, a chemical behavior similar to I- and I/S-type granitoidsof the Famatinian magmatic arc (Fig. 14A–D). The higher K/Na ratios ex-hibit by the studied granitoids in relation to basaltic-amphibolite-derivedmelts (Fig. 14D) could be attributed to re-melting of older calc-alkalinerocks that were themselves formed by mantle–crust interactions, as pos-tulated by Patiño Douce (1999) and Grosse et al. (2011).
Fig. 13. A–B diagram of Debon and Le Fort (1983) for the studied northernmost Rio Docearc rocks. Granitoid compositional fields in blue are from Villaseca et al. (1998). Dashedred arrows are trends of experimental melts from different protoliths: 1 = pelite-derivedmelt (Vielzeuf and Holloway, 1988); 2 = greywacke-derived melt (Conrad et al., 1988);3 = amphibolite-derived melt (Beard and Lofgren, 1991). Sample symbols are as inFigs. 4 and 5.
Given the regional scenario, we propose that the origin of thestudied granitic rocks involved mixing of both dehydration-melting of meta-igneous (amphibolites) rocks and dehydration-melting of metasedimentary (greywackes) rocks. The real contribu-tion of each end-member is, however, a challenging work still to bedone. In this context, the involvement of mantle-related magmasseems to be effective as triggering mechanism for the crustal melt-ing process.
5.1. Tectonic implications
In the previous section we made a case for significant recycling ofcontinental crust at the termination of the Rio Doce magmatic arc. Theacross-arc variation from I- to S-type granites observed in theFamatinianmagmatic arc, led Grosse et al. (2011) to interpret a progres-sive increase in the contribution of metasedimentary material towardthe interior of the continent, possibly at progressively higher crustallevels and associated with increasing crustal thickness. They argued infavor of a single and continuous subduction-related setting, in whichmetaluminous magmas are produced at continental margins, whileperaluminous magmas (with minor assimilation of mafic melts) aregenerated in the continental interior via crustal recycling and substan-tial melting of metasedimentary rocks. Considering that the Araçuaí–West Congo orogen developed in an uncommon confined environment,involving the closure of a basin that was partially ensialic and partiallyfloored by oceanic crust (Pedrosa-Soares et al., 1992, 1998, 2001,2008; Alkmim et al., 2006; Queiroga et al., 2007), it is likely that the ter-minal segment of the Rio Doce arc developed in the almost fully ensialicportion of the basin, with minor influence of oceanic crust subductionand mafic mantle-derived magmas. In this scenario, the increase inthe contribution of metasedimentary material must have occurredfrom south to north, as indicated by the higher peraluminous and tran-sitional I/S-type character of the studied G1 Supersuite granitoids.The tectonic panorama of northern Rio Doce arc is illustrated bythe block-diagram of Fig. 15. Nevertheless, participation ofmetasedimentary material was not sufficient to erase the arc- andsubduction-related chemical signatures (e.g., Niu et al., 2013) ofthe investigated granitoids (e.g., Fig. 7).
6. Conclusions
Field and petrographic observations, along with the results of ourgeochemical, isotopic and geochronological study of the granitoids
Fig. 14.Major element diagramsmodified from Patiño Douce (1999) and Grosse et al. (2011), with compositions of the studied Rio Doce arc rocks and of Famatinian arc granitoids (fromGrosse et al., 2011). Fields are compositions of experimental dehydration-melting of different types of rocks (for source of data see Patiño Douce, 1999).
456 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
forming the northernmost and almost fully intracontinental segment ofthe Rio Doce magmatic arc lead to the following conclusions:
1. The terminal segment of the arc consists of partially deformed andmigmatized granodiorites, tonalites, and minor monzogranites.
2. Differently from the hornblende–biotite bearing I-type granites(ASI = 1.02) of the central and southern portions of the arc, thenorthernmost G1 Supersuite granitoids in general do not containhornblende, are poor in mafic enclaves and exhibit characteristics(e.g., average ASI = 1.07) of both I-type and transitional I/S-typegranites (sensu Grosse et al., 2011).
3. Chemically, the granitoids correspond to magnesian, slightlyperaluminous, calcic- to calc-alkaline, medium- to high-K acid-silicic rocks, in this respect similar to Cordilleran-type granites de-scribed worldwide.
4. The obtained U–Pb zircon ages indicate that terminal branch of thearc developed between 618 Ma and 574 Ma and underwent partialmelting between 589 Ma and 555 Ma, in the course of the regionalmetamorphic event affecting the Araçuaí orogen. Ages obtained inzircon cores suggest significant contribution of the basement and Je-quitinhonha Complex rocks in the generation of the northern seg-ment of the arc.
5. The Sr and Nd isotopic composition of the granitoids, combined withLu–Hf data also point toward a strong involvement of the continentalcrust in the generation of studied granitoids.
6. Partial melting and mixing of metasedimentary and meta-igneous(amphibolites) rocks were probably the main process involved inthe generation of the arc terminal branch. However, further workin key places is required for the evaluation of the crustal and mantlesourcing in the production of the G1 Supersuite of the northern RioDoce arc.
Fig. 15. Block-diagram showing the tectonic setting of the Rio Docemagmatic arc and its terminal segment in the continental domain of the Macaúbas basin, whose closure led to devel-opment of the Araçuaí orogen.
457L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.gr.2015.07.015.
Acknowledgments
We are grateful to the Coordenação de Aperfeiçoamento de Pessoalde Nível Superior (CAPES) and Conselho Nacional de DesenvolvimentoCientífico e Tecnológico (CNPq) for the financial support. We wouldlike to thank Allan Collins for the editorial handling and an anonymousreviewer for the careful corrections and suggestions that greatly im-proved the manuscript. LG thanks the Prof. Edward Sawyer for helpfulcomments about migmatitic features. The technical staffs of theCPGeo — University of São Paulo (Maurício de Souza, Vasco dos Loios,and Walter Sproesser), LAGIR — State University of Rio de Janeiro(Carla Neto, Gilberto Vaz, João Barcellos, and Ricardo Travassos), aswell as of the NanoLab/RedeMat (Ney Sampaio) are thanked forlaboratorial facilities.
References
Alkmim, F.F., Marshak, S., Fonseca, M.A., 2001. Assembling West Gondwana in theNeoproterozoic: clues from the São Francisco Craton region, Brazil. Geology 29,319–322.
Alkmim, F.F., Marshak, S., Pedrosa-Soares, A.C., Peres, G.G., Cruz, S.C.P., Whittington, A.,2006. Kinematic evolution of the Araçuaí–West Congo orogen in Brazil and Africa:nutcracker tectonics during the Neoproterozoic assembly of Gondwana. PrecambrianResearch 149, 43–64.
Almeida, F.F.M., Litwinski, N., 1984. Província Mantiqueira: setor setentrional. In: deAlmeida, F.F.M., Hasui, Y. (Eds.), O Pré-Cambriano do Brasil. Editora Edgar Blücher,São Paulo, pp. 282–307.
Basei, M.A.S., Brito Neves, B.B., Siga-Junior, O., Babinski, M., Pimentel, M.M., Tassinari,C.C.G., Hollanda, M.H.B., Nutman, A., Cordani, U.G., 2010. Contribution of SHRIMPU–Pb zircon geochronology to unravelling the evolution of Brazilian Neoproterozoicfold belts. Precambrian Research 183, 112–144.
Beard, J.S., Lofgren, G.E., 1991. Dehydration melting and water-saturatedmelting of basal-tic and andesitic greenstones and amphibolites at 1, 3, and 6.9 kbar. Journal of Petrol-ogy 32, 365–401.
Brito Neves, B.B., Campos Neto, M.C., Fuck, R., 1999. From Rodinia to Western Gond-wana: an approach to the Brasiliano/Pan-African cycle and orogenic collage. Ep-isodes 22, 155–199.
Campos-Neto, M.C., 2000. Orogenic systems from southwestern Gondwana. An approachto Brasiliano–Pan African Cycle and orogenic collage in south eastern Brazil. In:Cordani, U.G., Milani, E.J., Thomaz-filho, A., Campos, D.A. (Eds.), Tectonic Evolutionof South America. 31st International Geological Congress, Rio de Janeiro, pp. 335–365.
Carvalho, J.B., Pereira, L.M.M., 2000. Petrografia, Relatório Integrado Etapa 1, folhas SE.24-V, SE.23-Z, SE.24-Y (partes), Estado de Minas Gerais, Escala 1:500,000 (93 p.). In:Projeto Leste, CPRM-CODEMIG (Eds.), Belo Horizonte, pp. 335–365.
Chappell, B.W., White, A.J.R., 1974. Two contrasting granite types. Pacific Geology 8,173–174.
Chappell, B.W., White, A.J.R., 2001. Two contrasting granite types: 25 years later.Australian Journal of Earth Sciences 48, 489–499.
Chappell, B.W., Colleen, J.B., Doone, W., 2012. Peraluminous I-type granites. Lithos 153,142–153.
Chemale Jr., F., Kawashita, K., Dussin, I.A., Ávila, J.N., Justino, D., Bertotti, A., 2012. U–Pb zir-con in situ dating with LA–MC–ICP–MS using a mixed detector configuration. Anaisda Academia Brasileira de Ciências 84, 275–295.
Chu, N.C., Taylor, R.N., Chavagnac, V., Nesbitt, R.W., Boella, R.M., Milton, J.A., German, C.R.,Bayon, G., Burton, K., 2002. Hf isotope ratio analysis using multi-collector inductivelycoupled plasma mass spectrometry: an evaluation of isobaric interference correc-tions. Journal of Analytical Atomic Spectrometry 17, 1567–1574.
Conrad, W.K., Nicholls, I.A., Wall, V.J., 1988. Water-saturated and undersaturated meltingof metaluminous and peraluminous crustal composition at 10 kb: evidence for theorigin of silicic magmas in the Taupo Volcanic Zone, New Zealand, and other occur-rences. Journal of Petrology 29, 765–803.
Cordani, U.G., D'Agrella-Filho, M.S., Brito Neves, B.B., Trindade, R.I.F., 2003. Tearing upRodinia: the Neoproterozoic paleogeography of South American cratonic fragments.Terra Nova 15, 350–359.
De Campos, C., Mendes, J., Ludka, I., de Medeiros, S., de Moura, J., Wallfass, C., 2004. A re-view of the Brasiliano magmatism in southern Espírito Santo, Brazil, with emphasison post-collisional magmatism. In: Weinberg, R., Trouw, R., Fuck, R., Hackspacher,P. (Eds.), The 750–550 Ma Brasiliano Event of South America. Journal of the VirtualExplorer, Electronic Edition1441-8142 vol. 17 (Paper 1).
De la Roche, H., Leterrier, J., Grande Claude, P., Marchal, M., 1980. A classification of volca-nic and plutonics rocks using R1–R2 diagrams and major element analyses — its re-lationships and current nomenclature. Chemical Geology 29, 183–210.
Debon, F., Le Fort, P., 1983. A chemical–mineralogical classification of common plutonicrocks and associations. Transactions of the Royal Society of Edinburgh: Earth Sciences73, 135–149.
DePaolo, D.J., 1981. Neodymium isotopes in the Colorado Front Range and implicationsfor crust formation and mantle evolution in the Proterozoic. Nature 291, 193–197.
Dunphy, J.M., Ludden, J.N., 1998. Petrological and geochemical characteristics of aPaleoproterozoic magmatic arc (Narsajuaq terrane, Ungava orogen, Canada) andcomparisons to Superior Province granitoids. Precambrian Research 91, 109–142.
Figueiredo, M.C.H., Campos Neto, M.C., 1993. Geochemistry of the Rio Docemagmatic arc,southeastern Brazil. Anais da Academia Brasileira de Ciências 65, 63–81.
Fontes, C.Q., 2000. Folha Águas Formosas. Projeto Leste. CPRM-CODEMIG. http://www.portalgeologia.com.br/mapa/.
Frost, B.R., Arculus, R.J., Barnes, C.G., Collins, W.J., Ellis, D.J., Frost, C.D., 2001. A geochemicalclassification of granitic rocks. Journal of Petrology 42, 2033–2048.
Gomes, A. C. B., 2008. Projeto Jequitinhonha: Folha Rio do Prado (SE.24-V-A-VI), 1:100,000. Programa Geologia do Brasil. Belo Horizonte, CPRM.
Gonçalves, L.E. da S., Alkmim, F.F., Pedrosa-Soares, A.C., 2010. Características geoquímicasda Suíte G1, arco magmático do Orógeno Araçuaí, entre Governador Valadares eIpanema, MG. Revista da Escola de Minas 63, 457–464.
Gonçalves, L., Farina, F., Lana, C., Pedrosa-Soares, A.C., Alkmim, F., Nalini Jr., H.A., 2014.New U–Pb ages and lithochemical attributes of the Ediacaran Rio Doce magmaticarc, Araçuaí confined orogen, southeastern Brazil. Journal of South American EarthSciences 52, 129–148.
Gonçalves-Dias, T., Pedrosa-Soares, A.C., Dussin, I.A., Alkmim, F.F., Caxito, F.A., Silva, L.C.,Noce, C.M., 2011. Maximum sedimentation age and provenance of the JequitinhonhaComplex in the type–area (Araçuaí orogen): first U–Pb (LA–ICP–MS) data from detri-tal zircon grains. Geonomos 19, 121–130.
Goscombe, B.D., Gray, D.R., 2008. Structure and strain variation at mid-crustal levelsin a transpressional orogen: a review of Kaoko Belt structure and the character ofWest Gondwana amalgamation and dispersal. Gondwana Research 13, 45–85.
Gradim, C., Roncato, J., Pedrosa-Soares, A.C., Cordani, U., Dussin, I., Alkmim, F.F., Queiroga,G., Jacobsohn, T., da Silva, L.C., Babinski, M., 2014. The hot back-arc zone of theAraçuaí orogen, Eastern Brazil: from sedimentation to granite generation. BrazilianJournal of Geology 44 (1), 155–180.
458 L. Gonçalves et al. / Gondwana Research 36 (2016) 439–458
Griffin, W.L., Pearson, N.J., Belousova, E.A., Saeed, A., 2006. Comment: Hf-isotope hetero-geneity in zircon 91500. Chemical Geology 233, 358–363.
Grosse, P., Bellos, L.I., de los Hoyos, C.R., Larrovere, M.A., Rossi, J.N., Toselli, A.J., 2011.Across-arc variation of the Famatinian magmatic arc (NW Argentina) exemplifiedby I-, S- and transitional I/S-type Early Ordovician granitoids of the Sierra de Velasco.J. S. Am. Earth Sci. 32, 110–126.
Heilbron, M., Tupinambá, M., Valeriano, C., Armstrong, R., Eirado-Silva, L.G., Melo, R.S.,Simonetti, A., Pedrosa-Soares, A.C., Machado, N., 2013. The Serra da Bolívia Complex:the record of a new Neoproterozoic arc-related unit at Ribeira belt. Precambrian Re-search 238, 158–175.
Jackson, S.E., Pearson, N.J., Griffin, W.L., Belousova, E.A., 2004. The application of laserablation–inductively coupled plasma–mass spectrometry to in situ U–Pb zircon geo-chronology. Chemical Geology 211, 47–69.
Kemp, A.I.S., Hawkesworth, C.H., 2003. Granitic perspectives on the generation and secu-lar evolution of the continental crust. In: Holland, H.D., Turekian, K.K. (Eds.), Treatiseon Geochemistry. Elsevier, Amsterdam, pp. 349–431.
Kemp, A.I.S., Hawkesworth, C.J., Foster, G.L., Paterson, B.A., Woodhead, J.D., Hergt, J.M.,Gray, C.M., Whitehouse, M.J., 2007. Magmatic and Crustal differentiation history ofgranitic rocks from Hf−O Isotopes in Zircon. Science 315, 980–983.
Kemp, A.I.S., Hawkesworth, C.J., Paterson, B.A., Kinny, P., 2006. Episodic growth of theGondwana Supercontinent from hafnium and oxygen isotopes in zircon. Nature439, 580–583.
Liu, R., Li, J.-W., Bi, S.-J., Hu, H., Chen, M., 2013. Magmamixing revealed from in situ zirconU–Pb–Hf isotope analysis of the Muhuguan granitoid pluton, eastern Qinling orogen,China: implications for late Mesozoic tectonic evolution. International Journal ofEarth Sciences (Geologische Rundschau) 102, 1583–1602.
Ludwig, K.R., 2003. Using Isoplot/Ex, version 3.00, a geochronological toolkit for MicrosoftExcel. Berkeley Geochronology Center, Special Publication 1 (43 pp.).
Maniar, P.D., Picolli, P.M., 1989. Tectonic discrimination of granitoids. Geological Society ofAmerica Bulletin 101, 635–643.
Martins, V.T. de S., Teixeira, W., Noce, C.M., Pedrosa-Soares, A.C., 2004. Sr and Nd charac-teristics of Brasiliano/Pan-Africano granitoid plutons of the Araçuaí orogen, south-eastern Brazil: tectonic implications. Gondwana Research 7, 75–89.
McCulloch, M.T., Chappell, B.W., 1982. Nd isotopic characteristics of S- and I-type granites.Earth and Planetary Science Letters 58, 51–64.
McDonough, W.F., Sun, S.-S., 1995. The composition of the Earth. Chemical Geology 120,223–253.
Mondou, M., Egydio-Silva, M., Vauchez, A., Raposo, M.I.B., Bruguier, O., Oliveira, A.F.,2012. Complex, 3D strain patterns in a synkinematic tonalite batholith from theAraçuaí Neoproterozoic orogen (Eastern Brazil): evidence from combined mag-netic and isotopic chronology studies. Journal of Structural Geology 39, 158–179.
Moreira, M.D., 2000. Folha Mucuri. Projeto Leste. CPRM-CODEMIG. http://www.portalgeologia.com.br/mapa/.
Morel, M.L.A., Nebel, O., Nebel-Jacobsen, Y.J., Miller, J.S., Vroon, P.Z., 2008. Hafnium isotopecharacterization of the GJ-1 zircon reference material by solution and laser-ablationMC–ICPMS. Chemical Geology 255, 231–235.
Nakamura, N., 1974. Determination of REE, Ba, Fe, Mg, Na and K in carbonaceous and or-dinary chondrites. Geochimica et Cosmochimica Acta 38, 757–775.
Nalini Jr., H.A., 1997. Caractérisation des suites magmatiques néoprotérozoiques de la re-gion de Conselheiro Pena et Galiléia (Minas Gerais, Brésil). Etude géochimique etstructurale des suites Galiléia et Urucum et relations avec les pegmatites à élémentsrares associées. Ecole des Mines de Saint Etienne et Ecole des Mines de Paris (Ph.D.Thesis, 237 pp.).
Nalini Jr., H.A., Bilal, E., Correia Neves, J.M., 2000. Syn-collisional peraluminousmagmatism in the Rio Doce region: mineralogy, geochemistry and isotopicdata of the Urucum suite (eastern Minas Gerais state, Brazil). Revista Brasileirade Geociencias 30, 120–125.
Nalini Jr., H.A., Machado, R., Bilal, E., 2005. Geoquímica e Petrogênese da SuíteGaliléia: Exemplo de Magmatismo Tipo-I Metaluminoso Pré-ColisionalNeoproterozóico da Região do Médio Vale do Rio Doce (MG). Revista Brasileirade Geociencias 35, 23–34.
Niu, Y., Zhao, Z., Zhuc, Di-C., Mo, X., 2013. Continental collision zones are primary sites fornet continental crust growth — a testable hypothesis. Earth-Science Reviews 127,96–110.
Noce, C.M., Pedrosa-Soares, A.C., Silva, L.C., Armstrong, R., Danielle Piuzana, D., 2007. Evo-lution of polycyclic basement complexes in the Araçuaí orogen, based on U–PbSHRIMP data: implications for Brazil–Africa links in Paleoproterozoic time. Precam-brian Research 159, 60–78.
Novo, T.A., Pedrosa-Soares, A.C., Noce, C.M., Alkmim, F.F., Dussin, I., 2010. Rochascharnockíticas do sudeste de Minas Gerais: a raiz granulítica do arco magmático doOrógeno Araçuaí. Revista Brasileira de Geociencias 40, 573–592.
Paes, V.J.C., 2000. Folha Teófilo Otoni. Projeto Leste. CPRM-CODEMIG. http://www.portalgeologia.com.br/mapa/.
Paes, V.J.C., Raposo, F.O., Pinto, C.P., Oliveira, F.A.R., 2010. Projeto Jequitinhonha, Estadosde Minas Gerais e Bahia: texto explicativo. Geologia e Recursos Minerais das FolhasComercinho, Jequitinhonha, Almenara, Itaobim, Joaíma e Rio do Prado, ProgramaGeologia do Brasil: Belo Horizonte, CPRM (376 pp.).
Pankhurst, R.J., Rapela, C.W., Saavedra, J., Baldo, E.G., Dahlquist, J.A., Pascua, I., Fanning,C.M., 1998. The Famatinian arc in the central Sierras Pampeanas: an early to mid-Ordovician continental arc on the Gondwana margin. In: Pankhurst, R.J., Rapela,C.W. (Eds.), The Proto-AndeanMargin of Gondwana. Geological Society, London, Spe-cial Publications 142, pp. 343–367.
Patchett, P.J., Tatsumoto, M., 1980. A routine high-precision method for Lu–Hf isotope geo-chemistry and chronology. Contributions to Mineralogy and Petrology 75, 263–267.
Patiño Douce, A.E., 1999. What do experiments tell us about the relative contributions ofcrust and mantle to the origin of granitic magmas? In: Castro, A., Fernández, C.,Vigneresse, J.L. (Eds.), Understanding Granites: Integrating New and Classical Tech-niques. Geological Society, London, Special Publications 168, pp. 55–75
Pedrosa-Soares, A.C., Wiedemann-Leonardos, C.M., 2000. Evolution of the Araçuaí beltand its connection to the Ribeira Belt, Eastern Brazil. In: Cordani, U.G., Milani, E.J.,Thomaz Filho, A., Campos, D.A. (Eds.), Tectonic Evolution of South America. Interna-tional Geological Congress, Rio de Janeiro, pp. 265–285.
Pedrosa-Soares, A.C., Noce, C.M., Vidal, P., Monteiro, R.L.B.P., Leonardos, O.H., 1992. To-ward a new tectonic model for the Late Proterozoic Araçuaí (SE Brazil)–WestCongolian (SW Africa) Belt. Journal of South American Earth Sciences 6, 33–47.
Pedrosa-Soares, A.C., Vidal, P., Leonardos, O.H., Brito Neves, B.B., 1998. Neoproterozoicoceanic remnants in eastern Brazil: further evidence and refutation of an exclusivelyensialic evolution for the Araçuaí–West Congo orogen. Geology 26, 519–522.
Pedrosa-Soares, A.C., Noce, C.M., Wiedemann, C.M., Pinto, C.P., 2001. The Araçuaí–WestCongo orogen in Brazil: an overview of a confined orogen formed during Gondwana-land assembly. Precambrian Research 110, 307–323.
Pedrosa-Soares, A.C., Alkmim, F.F., Tack, L., Noce, C.M., Babinski, M., Silva, L.C., Martins-Neto, M.A., 2008. Similarities and differences between the Brazilian and African coun-terparts of the Neoproterozoic Araçuaí–West-Congo orogen. Geological Society,London, Special Publications 294, 153–172.
Pedrosa-Soares, A.C., De Campos, C., Noce, C.M., Silva, L.C., Novo, T., Roncato, J., Medeiros,S., Castañeda, C., Queiroga, G., Dantas, E., Dussin, I., Alkmim, F.F., 2011. LateNeoproterozoic–Cambrian granitic magmatism in the Araçuaí orogen (Brazil), theEastern Brazilian Pegmatite Province and related mineral resources. Geological Soci-ety, London, Special Publications 350, 25–51.
Peixoto, E., Pedrosa-Soares, A.C., Alkmim, F.F., Dussin, I.A., 2015. A suture-related accre-tionary wedge formed in the Neoproterozoic Araçuaí orogen (SE Brazil) duringWest-ern Gondwanaland assembly. Gondwana Research 27 (2), 878–896.
Porada, H., 1989. Pan-African rifting and orogenesis in southern to equatorial Africa andEastern Brazil. Precambrian Research 44, 103–136.
Queiroga, G.N., Pedrosa-Soares, A.C., Noce, C.M., Alkmim, F.F., Pimentel, M.M., Dantas, E.,Martins, M., Castañeda, C., Suita, M.T.F., Prichard, H., 2007. Age of the Ribeirão daFolha ophiolite, Araçuaí orogen: the U–Pb zircon dating of a plagiogranite. Geonomos15, 61–65.
Rickwood, P.C., 1989. Boundary lines within petrologic diagrams which use oxides ofmajor and minor elements. Lithos 22, 247–263.
Rudnick, R., 1995. Making continental crust. Nature 378, 571–578.Sampaio, A.R., 2000. Folha Padre Paraíso. Projeto Leste. CPRM-CODEMIG. http://www.
portalgeologia.com.br/mapa/.Silva, L.C., Armstrong, R., Noce, C.M., Carneiro, M., Pimentel, M., Pedrosa-Soares, A.C., Leite,
C., Vieira, V.S., Silva, M., Paes, V., Cardoso-Filho, J., 2002. Reavaliação da evoluçãogeológica em terrenos pré-cambrianos brasileiros com base em novos dados U–PbSHRIMP, parte II: Orógeno Araçuaí, Cinturão Móvel Mineiro e Cráton São FranciscoMeridional. Revista Brasileira de Geociencias 32, 513–528.
Silva, L.C., Pedrosa-Soares, A.C., Armstrong, R., Noce, C.M., 2011. Determinando a duraçãodo período colisional do Orógeno Araçuaí com base em geocronologia U–Pb de altaresolução em zircão: uma contribuição para a história da amalgamação do GondwanaOcidental. Geonomos 19, 180–197.
Sláma, J., Košler, J., Condon, D.J., Crowley, J.L., Gerdes, A., Hanchar, J.M., Horstwood, M.S.A.,Morris, G.A., Nasdala, L., Norberg, N., Schaltegger, U., Schoene, B., Tubrett, M.N.,Whitehouse, M.J., 2008. Plešovice zircon — a new natural reference material for U–Pb and Hf isotopic microanalysis. Chemical Geology 249, 1–35.
Sun, S.S., McDonough, W.E., 1989. Chemical and isotopic systematics of oceanic basalts:implications for mantle composition and processes. In: Saunder, A.D., Norry, J.M.(Eds.), Magmatism in Ocean Basins. Geological Society, London, Special Publications42, pp. 313–345.
Tanaka, T., Togashi, S., Kamioca, H., 2000. JNd-1: a neodymium isotopic reference in con-sistency with La Jolla neodymium. Chemical Geology 168, 279–281.
Tedeschi, M., 2013. Caracterização do arco magmático do Orógeno Araçuaí entre FreiInocêncio e Itambacuri, MG (Master Thesis), UFMG, Belo Horizonte, Brazil (127 pp.).
Torquato, J.R., Cordani, U.G., 1981. Brazil–Africa geological links. Earth-Science Reviews17, 155–176.
Tupinambá, M., Heilbron, M., Valeriano, C., Porto Jr., R., de Dios, F.B., Machado, N., Silva,L.G.E., Almeida, J.C.H., 2012. Juvenile contribution of the Neoproterozoic Rio NegroMagmatic Arc (Ribeira Belt, Brazil): implications for Western Gondwana amalgam-ation. Gondwana Research 21, 422–438.
Vieira, V.S., 2007. Significado do Grupo Rio Doce no Contexto do Orógeno Araçuaí (Ph.D.Thesis), IGC-UFMG, Belo Horizonte, Brasil (117 pp.).
Vielzeuf, D., Holloway, J.R., 1988. Experimental determination of the fluid-absent meltingrelations in the pelitic system. Consequences for crustal differentiation. Contributionsto Mineralogy and Petrology 98, 257–276.
Villaros, A., Buick, I.S., Stevens, G., 2012. Isotopic variations in S−type granites: an inher-itance from a heterogeneous source? Contrib. Mineral. Petrol. 163 (2), 243–257.
Villaseca, C., Barbero, L., Herreros, V., 1998. A re-examination of the typology ofperaluminous granite types in intracontinental orogenic belts. Transactions of theRoyal Society of Edinburgh: Earth Sciences 89, 113–119.
Wu, F.Y., Yang, Y.H., Xie, L.W., Yang, J.H., Xu, P., 2006. Hf isotopic compositions of the stan-dard zircons and baddeleyites used in U–Pb geochronology. Chemical Geology 234,105–126.
Top Related